diff --git a/tools/Python/mctest/mctest.py b/tools/Python/mctest/mctest.py
old mode 100755
new mode 100644
index 831a36109..47b9ac251
--- a/tools/Python/mctest/mctest.py
+++ b/tools/Python/mctest/mctest.py
@@ -16,6 +16,7 @@
 import shutil
 import platform
 import subprocess
+import io
 
 sys.path.append(os.path.join(os.path.dirname(__file__), '..'))
 from mccodelib import utils, mccode_config
@@ -195,6 +196,34 @@ def extract_testvals(datafolder, monitorname):
                 return (I, I_err, N)
                 break
 
+def parse_detector_I_value(resfile_path, detector_name):
+    """
+    Return (value_float, success_bool, raw_value_str_or_None).
+    value_float is the parsed float (or -1.0 on failure).
+    success_bool is True when a value was parsed successfully.
+    raw_value_str_or_None is the string extracted (before conversion) or None.
+    """
+    prefix = f"Detector: {detector_name}_I="
+    try:
+        with open(resfile_path, "r", encoding="utf-8", errors="replace") as f:
+            for line in f:
+                if line.startswith(prefix):
+                    # extract after first '=' then take up to first whitespace
+                    # matches the shell pipeline: cut -f2 -d= | cut -f1 -d' '
+                    _, _, after_eq = line.partition("=")
+                    raw = after_eq.split()[0] if after_eq else ""
+                    if raw == "":
+                        return -1.0, False, None
+                    try:
+                        return float(raw), True, raw
+                    except ValueError:
+                        return -1.0, False, raw
+        return -1.0, False, None
+    except FileNotFoundError:
+        return -1.0, False, None
+    except OSError:
+        return -1.0, False, None
+
 def mccode_test(branchdir, testdir, limitinstrs=None, instrfilter=None, compfilter=None, version=None):
     ''' this main test function tests the given mccode branch/version '''
     skipped=False
@@ -394,6 +423,7 @@ def mccode_test(branchdir, testdir, limitinstrs=None, instrfilter=None, compfilt
             failed=True
             continue
 
+        resbase="(No file)"
         # test value extraction
         if not didwrite_nexus:
             extraction = extract_testvals(join(testdir, test.instrname, str(test.testnb)), test.detector)
@@ -402,26 +432,36 @@ def mccode_test(branchdir, testdir, limitinstrs=None, instrfilter=None, compfilt
             else:
                 test.testval = -1
                 failed=True
+            resbase ="run_stdout_%d.txt" % (test.testnb)
+            resfile = join(testdir,test.instrname,resbase)
         # Look for detector output in run_stdout
         else:
             metalog = LineLogger()
-            resfile = join(testdir,test.instrname,"run_stdout_%d.txt" % (test.testnb))
-            cmd = r"grep %s_I= %s | head -1 | cut -f2 -d= | cut -f1 -d' '" %(test.detector, resfile)
-            utils.run_subtool_to_completion(cmd, stdout_cb=metalog.logline)
-            try:
-                test.testval=float(metalog.lst[0])
-            except:
+            resbase ="run_stdout_%d.txt" % (test.testnb)
+            resfile = join(testdir,test.instrname,resbase)
+            val, ok, raw = parse_detector_I_value(resfile, test.detector)
+            if ok:
+                test.testval = val
+            else:
                 test.testval=-1
                 failed=True
 
-        if test.didrun and not failed:
+        percent=0
+        if test.didrun:
+            if failed:
+                suffix += " + !! RUNTIME FAILURE - see %s !! " % (resbase)
             formatstr = "%-" + "%ds: " % (maxnamelen+1) + \
                 "{:3d}.".format(math.floor(test.runtime)) + str(test.runtime-int(test.runtime)).split('.')[1][:2]
             if test.targetval!=0: # Normal situation, non-zero target value
-                logging.info(formatstr % test.get_display_name() + "    [val: " + str(test.testval) + " / " + str(test.targetval) + " = " + str(round(100.0*test.testval/test.targetval)) + " %]" + suffix)
+                percent=round(100.0*test.testval/test.targetval)
+                if percent<80 or percent>120:
+                    suffix += " <--- BIG DISCREPANCY??"
+                logging.info(formatstr % test.get_display_name() + "    [val: " + str(test.testval) + " / " + str(test.targetval) + " = " + str(percent) + " %]" + suffix)
             else:                 # Special case, expected test target value is 0
                 logging.info(formatstr % test.get_display_name() + "    [val: " + str(test.testval) + " vs " + str(test.targetval) + " (absolute vs 0) ]" + suffix)
-
+        else:
+            logging.info((formatstr % test.get_display_name()) + (" !! [TEST INDICATES RUNTIME ERROR - see %s  + suffix ] !!" % (resbase)))
+        suffix=""
         # save test result to disk
         test.testcomplete = True
         if not skipped: